A hybrid speech recognition model based on HMM and fuzzy PPM

Abstract

Hidden Markov model (HMM) is a robust statistical methodology for automatic speech recognition. It has been tested in a wide range of applications. A prediction approach traditionally applied for the text compression and coding. Prediction by partial matching (PPM) which is a finite-context statistical modeling technique and can predict the next characters based on the context has shown a great potential in developing novel solutions to several language modeling problems in speech recognition. These two different approaches have their own spatial features respectively contributing to speech recognition. However, no work has been reported in integrating them at an attempt to form a hybrid speech recognition scheme. Taking the advantages of these two approaches, we propose a hybrid speech recognition model based on HMM and fuzzy PPM. The competitive and promising performance of the approach in speech recognition has been demonstrated by an experiment.